Carl Boettiger

Sweave is some­thing of a gold stan­dard in repro­ducible research. It cre­ates a dynamic doc­u­ment, writ­ten in a mix of LaTeX and R code where the results of the analy­sis (num­bers, fig­ures, tables) are auto­mat­i­cally gen­er­ated from the code and inserted into the result­ing pdf doc­u­ment, mak­ing them easy to update if the data or …

I have recently mod­i­fied the basic work­flow of my lab note­book since dis­cov­er­ing knitr. Before, I would write code files which I could track on github, push fig­ures cre­ated by the code to flickr, and then write a note­book entry on word­press describ­ing what I was doing. I’d embed each fig­ure I wanted into the …

I’m not always care­ful in cit­ing all the R pack­ages I use. R actu­ally has some rather nice built-in mech­a­nisms to sup­port this, so I really have no excuse. Here’s some quick exam­ples: To cite the ouch pack­age in pub­li­ca­tions use: Aaron A. King and Mar­guerite A. But­ler (2009), ouch: Ornstein-Uhlenbeck mod­els for phy­lo­ge­netic com­par­a­tive …

I’ve cre­ated an R pack­age that uses social media tools for repro­ducible research.  The goal of the pack­age is this: when­ever I run a code, out­put fig­ures are auto­mat­i­cally added to my fig­ure repos­i­tory (Flickr), linked to the time­stamped ver­sion of the code that pro­duced them in the code repos­i­tory.  Fig­ures should be tagged by …